tarekziade HF Staff
Restore full contract and slim-model descriptions on the landing page
dcf1835 verified | tags: | |
| - mic | |
| - test-artifact | |
| base_model: stepfun-ai/GOT-OCR-2.0-hf | |
| # got_ocr2-mic | |
| Slim model weights, executable usage and framework expectations in one repository. | |
| - [Contract manifest](integration.json) | |
| - [Transformers usage](integrations/transformers/usage.md) | |
| - [Transformers expectations](integrations/transformers/expectations) | |
| - [Fixture provenance](build.json) | |
| ## Contract | |
| ### GOT-OCR2 | |
| Runnable usage and the framework expectations directory are linked above. | |
| ## Slim model | |
| ### Slim fixture: mic-got_ocr2-slim | |
| A 13.0M-parameter fixture for the [got_ocr2 MIC](https://huggingface.co/huggingface-mic/got_ocr2-mic), covering [`stepfun-ai/GOT-OCR-2.0-hf`](https://huggingface.co/stepfun-ai/GOT-OCR-2.0-hf). It is a test artifact, not a usable model, tagged `test-artifact` to keep it out of candidate-model search. | |
| - **Fixture key `6141fa8be862`.** It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking. | |
| - **Same architecture, smaller config:** `GotOcr2ForConditionalGeneration`, shrunk by the generic rule in [`sweep.py`](https://github.com/huggingface/transformers-mic/blob/main/experiments/sweep.py) (vision num_hidden_layers 2, text num_hidden_layers 2). | |
| - **Processor files copied byte for byte** from `stepfun-ai/GOT-OCR-2.0-hf`: `generation_config.json`, `preprocessor_config.json`, `special_tokens_map.json`, `tokenizer.json`, `tokenizer_config.json`. | |
| - **Overfit to the contract's training step** (250 steps). Random weights failed the input-sensitivity gate: `pixel_values:roll_batch` moved the logits by only 0 (threshold 0.001). | |
| - **Same tensor names as the real checkpoint.** The real checkpoint has every parameter, and so does the fixture. | |
| - **Saved in bfloat16**, the dtype `stepfun-ai/GOT-OCR-2.0-hf` loads in by default. | |
| - **Finite in float32, bfloat16, and float16** (forward, loss, gradients). | |
| Built by [`build_fixtures.py`](https://github.com/huggingface/transformers-mic/blob/main/example/build_fixtures.py); [`build.json`](build.json) records the pinned sources, gate results, and file hashes. This repository includes the slim weights, usage and framework expectations; rebuild the other files with `python example/build_fixtures.py --fixture mic-got_ocr2-slim`. | |